Haibo Xing

Haibo Xing

orcid

Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
13 July 2025
Verified
Conference Paper
Open access
HDHao Deng
Hao Deng
HXHaibo Xing
Haibo Xing
KMKanefumi Matsuyama
Kanefumi Matsuyama
YHYulei Huang
Yulei Huang
JHJinxin Hu
Jinxin Hu
JZJing Zhang
Jing Zhang


22 April 2025
Verified
Conference Paper
Open access
HXHaibo Xing
Haibo Xing
KMKanefumi Matsuyama
Kanefumi Matsuyama
HDHao Deng
Hao Deng
JHJinxin Hu
Jinxin Hu
YZYu Zhang
Yu Zhang
XZXiaoyi Zeng
Xiaoyi Zeng

13 July 2025
Verified
Conference Paper
Open access
HDHao Deng
Hao Deng
HXHaibo Xing
Haibo Xing
KMKanefumi Matsuyama
Kanefumi Matsuyama
MZMoyu Zhang
Moyu Zhang
JHJinxin Hu
Jinxin Hu
JZJing Zhang
Jing Zhang

16 October 2025 in arXiv
Verified
Preprint
Open access
PDF file
LMLingyu Mu
Lingyu Mu
HDHao Deng
Hao Deng
HXHaibo Xing
Haibo Xing
KLKaican Lin
Kaican Lin
ZZZhitong Zhu
Zhitong Zhu
JHJinxin Hu
Jinxin Hu
The integration of large language models (LLMs) into recommendation systems has revealed promising potential through their capacity to extract world knowledge for enhanced reasoning capabilities. However, current methodologies that adopt static schema-based prompting mechanisms encounter significant limitations: (1) they employ universal template structures that neglect the multi-faceted nature of user preference diversity; (2) they implement superficial alignment between semantic knowledge representations and behavioral feature spaces without achieving comprehensive latent space integration. To address these challenges, we introduce CoCo, an end-to-end framework that dynamically constructs user-specific contextual knowledge embeddings through a dual-mechanism approach. Our method realizes profound integration of semantic and behavioral latent dimensions via adaptive knowledge fusion and contradiction resolution modules. Experimental evaluations across diverse benchmark datasets and an enterprise-level e-commerce platform demonstrate CoCo's superiority, achieving a maximum 8.58% improvement over seven cutting-edge methods in recommendation accuracy. The framework's deployment on a production advertising system resulted in a 1.91% sales growth, validating its practical effectiveness. With its modular design and model-agnostic architecture, CoCo provides a versatile solution for next-generation recommendation systems requiring both knowledge-enhanced reasoning and personalized adaptation.

21 August 2025 in arXiv
Verified
Preprint
Open access
PDF file
YXYi Xu
Yi Xu
MZMoyu Zhang
Moyu Zhang
CLChenxuan Li
Chenxuan Li
Zhihao LiaoZhihao Liao
Zhihao Liao
HXHaibo Xing
Haibo Xing
JZJing Zhang
Jing Zhang
Recommender systems traditionally represent items using unique identifiers (ItemIDs), but this approach struggles with large, dynamic item corpora and sparse long-tail data, limiting scalability and generalization. Semantic IDs, derived from multimodal content such as text and images, offer a promising alternative by mapping items into a shared semantic space, enabling knowledge transfer and improving recommendations for new or rare items. However, existing methods face two key challenges: (1) balancing cross-modal synergy with modality-specific uniqueness, and (2) bridging the semantic-behavioral gap, where semantic representations may misalign with actual user preferences. To address these challenges, we propose Multimodal Mixture-of-Quantization (MMQ), a two-stage framework that trains a novel multimodal tokenizer. First, a shared-specific tokenizer leverages a multi-expert architecture with modality-specific and modality-shared experts, using orthogonal regularization to capture comprehensive multimodal information. Second, behavior-aware fine-tuning dynamically adapts semantic IDs to downstream recommendation objectives while preserving modality information through a multimodal reconstruction loss. Extensive offline experiments and online A/B tests demonstrate that MMQ effectively unifies multimodal synergy, specificity, and behavioral adaptation, providing a scalable and versatile solution for both generative retrieval and discriminative ranking tasks.